IPR, PGR, Or Reexamination? Choosing A Path To Challenge A Patent At The USPTO

The three ways to attack a patent’s validity at the Patent Office: what each one costs you, what each one buys you, and why the choice is harder than it was a few years ago

When a patent is asserted against your company, in a demand letter, a district court complaint, or an ITC investigation, one of the first strategic questions is whether to attack the patent itself, and if so, where. District court is one option. The Patent Office offers three separate paths, each with its own statute, timing rules, and consequences.

The path you pick determines what arguments you can make, whether you get to participate at all, what you give up in the district court case, how much it costs, and how quickly you get an answer. The calculus has also shifted since 2025, as the USPTO has changed how it decides which challenges to take up in the first place.

What is an inter partes review (IPR)?

An IPR is an adversarial trial proceeding before the Patent Trial and Appeal Board, governed by 35 U.S.C. §§ 311–319. Any person other than the patent owner may petition.

Three features define it.

1. Narrow grounds. An IPR can challenge claims only under § 102 (anticipation) or § 103 (obviousness), and only on the basis of patents and printed publications. Section 101 eligibility, § 112 indefiniteness, prior public use, and on-sale-bar arguments are off the table.
2. Hard timing rules. For a patent subject to the AIA’s first-inventor-to-file provisions, a petition cannot be filed until nine months after the patent issues, or, if a post-grant review was instituted, until that proceeding ends (§ 311(c)). More important for a defendant, § 315(b) bars a petition filed more than one year after service of a complaint alleging infringement of that patent. That one-year clock is the most common statutory way companies lose the IPR option. It runs while the business is still deciding whether to fight. Discretionary denial, discussed below, is now the other.
3. Estoppel. If the Board issues a final written decision as to a claim, § 315(e) estops the petitioner from later asserting, in district court, at the ITC, or back at the Office, any ground it raised or reasonably could have raised in the IPR as to that claim. The estoppel is claim-specific. Because an IPR can only raise patents and printed publications under §§ 102 and 103, prior public use or on-sale system art is not necessarily foreclosed merely because a similar technical disclosure appeared in a printed publication used in the IPR. Estoppel is still the real price of admission. Confirm its exact scope with counsel before you file.

In exchange, you get a full trial: expert declarations, depositions, a hearing, and a written decision, with a statutory deadline of one year from institution.

What is a post-grant review (PGR), and why is it rarely available?

A PGR, under 35 U.S.C. §§ 321–329, is the broadest tool and the hardest to use, because of a nine-month window.

The grounds are wide. Under § 321(b), a PGR can raise essentially any invalidity ground available as a defense under § 282(b)(2) or (3): § 101 subject-matter eligibility, § 112 written description and indefiniteness, prior public use and on-sale activity, in addition to § 102 and § 103. Best-mode failure is the notable carve-out. Where the strongest attack on a patent is eligibility or enablement, that breadth matters.

The catch is § 321(c): a PGR petition must be filed within nine months of the patent’s grant. It also applies only to patents whose claims have an effective filing date on or after March 16, 2013, the AIA’s first-inventor-to-file transition.

The practical consequence is simple. A PGR is almost never available to a company that first learns of a patent when it gets sued. It is available to a company watching a competitor’s applications and prepared to move the moment a patent issues. That is a portfolio-monitoring discipline, not a litigation reflex, and one reason we encourage clients in crowded spaces to track competitor filings rather than wait to be surprised.

What is an ex parte reexamination, and why is it back in the conversation?

Ex parte reexamination is the oldest of the three, under 35 U.S.C. §§ 301–307. Any person may request it at any time during the patent’s enforceability, and the Office orders reexamination if the request raises a substantial new question of patentability based on patents or printed publications.

Its defining trait is right there in the name. After the initial request, and one optional reply if the patent owner files a statement, the third-party requester is out. The examination proceeds between the examiner and the patent owner, who can amend claims. No discovery, no hearing, no seat at the table.

That is often a disadvantage. Reexamination still carries three offsetting advantages.

First, there is no IPR-style estoppel. Section 315(e) estoppel attaches to an IPR final written decision, not to a reexamination request. A reexam does not spend your district-court invalidity arguments (though they may be compromised if unsuccessful in the reexamination).

Second, there is no § 315(b) one-year bar. Reexamination remains available after the IPR window has closed.

Third, anonymity has historically been available. A reexamination request has not required the real-party-in-interest statement an IPR petition demands. Treat this one as in flux, though. In February 2026 the Office tightened what an anonymous request must certify when the patent was previously in an IPR or PGR. In July 2026 it proposed requiring identification of all real-parties-in-interest. Confirm the current rule before relying on anonymity.

One important trap: the estoppel runs the other way. If you have already lost an IPR on the merits, § 315(e)(1) estops you from then asking the Office to reexamine on grounds you raised or reasonably could have raised. You cannot use reexamination as a second bite after an IPR final written decision. A reexamination request already requires a certification that this estoppel does not bar it.

Has the Patent Office changed how it decides whether to take these cases?

Yes. This is the part that most reshaped the analysis.

On March 26, 2025, the USPTO issued a memorandum on interim processes for PTAB workload management that bifurcated the institution decision. Discretionary considerations are briefed and decided separately from the merits. A patent owner seeking discretionary denial files a dedicated brief within two months of the Notice of Filing Date Accorded. The petitioner opposes within one month after that brief is filed. The Office’s published guidance lists a non-exhaustive set of considerations, including Fintiv, General Plastic, Advanced Bionics, the strength of the unpatentability challenge, the extent of reliance on expert testimony, and the settled expectations of the parties, such as how long the claims have been in force.

Under that March 2025 memo, the Director decided the discretionary question and, if denial was not appropriate, referred the petition to a three-judge panel for the merits.

Then, on October 17, 2025, effective October 20, the Director withdrew the delegation of merits-based institution authority to PTAB panels. The Director now decides institution, discretionary and merits alike, typically by summary notice, in consultation with at least three PTAB judges. A proposed rule published the same day would tighten institution further still, including a mandatory stipulation broader than Sotera that would waive all § 102 and § 103 grounds in other forums. It had not issued as a final rule when this was written.

On March 11, 2026, the Office added further discretionary considerations relating to U.S. manufacturing and small-business use of AIA proceedings.

Two takeaways follow. First, an IPR petition is now two arguments, not one: the merits, and a separate argument for why the Office should reach the merits at all. A petition that is technically strong but discretionarily vulnerable (a long-issued patent, a fast-moving parallel district court case, a prior challenge to the same claims) may never be heard. A Sotera-style stipulation narrowing what you will argue elsewhere is one lever, and the Office weighs whether the stipulation materially reduces overlap.

Second, this is precisely why ex parte reexamination has moved from a niche tool back toward the center of validity strategy for many defendants. It is not a substitute for an IPR. You give up participation, and the Office has its own discretion over reexamination requests. It is a real path when the IPR door is closed or the discretionary risk is high.

This area is moving. The interim process, the discretionary considerations, who decides institution, and the rulemaking around it have each changed more than once, and more is pending. Confirm every procedural detail here with counsel against current Office guidance before you rely on it.

How does this interact with the district court case?

It should drive the district court case, not follow it. Three things I would flag.

The § 315(b) clock starts at service, not at the moment the business decides to take the case seriously. If an IPR is even possible, the invalidity search needs to start immediately.

A validity challenge is leverage, not just a defense. The prospect of the asserted claims being narrowed or cancelled changes the settlement math, particularly against an assertion built on a weak patent and a nuisance-value number. Specific settlement figures move constantly. Ask counsel for current data that fits your situation.

And the record you build early is the record you will need later. In Octane Fitness v. ICON Health & Fitness, a case for which I was the district-court trial counsel, and which we won on summary judgment before it reached the Supreme Court, a unanimous Court held that an “exceptional” case under 35 U.S.C. § 285 is simply one that “stands out from others with respect to the substantive strength of a party’s litigating position … or the unreasonable manner in which the case was litigated,” judged on the totality of the circumstances, and proved by a preponderance of the evidence rather than the clear-and-convincing standard courts had been applying. The prior art you develop for an invalidity challenge is often the same evidence showing a claim never should have been brought. Building the fee-shifting record starts on day one.

Working with counsel on a validity strategy

The estoppel and timing rules mean these decisions cannot sensibly be made by one team in isolation from the district court or ITC case sitting next to them. If a patent has been asserted against your company, or you are monitoring a competitor’s applications and want to be ready when one issues, the earlier the strategy is set, the more options you have.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.